The Structuring of Structure

Abstract

THERE is a somewhat dogmatic belief in certain schools of anthropology that those events which can be explained psychologically and those which can be explained structurally are mutually exclusive. Thus, it is often written that psychological variables are not suitable as explanatory devices for social events. Frequently those who believe that psychological explanation has no place in the social sciences do not support this view with logical or empirical evidence but merely refer their audience to the assertions of some eminent authority who has held similar views. For example, Needham (1962:4), who holds that psychological explanations are inapt to the solution of sociological problems supports this belief primarily by selectively quoting Durkheim's assertions on the subject. Service (1962:201) does much the same in a recent work intended as an advanced introduction to primitive social organization. These appeals to authority have no place in any work that aspires to scientific standards of validity. The canonization of the method and theory of the pioneers in social science can only lead to sterile repetitions of their work. The purpose of scholarship is not to ritualize, but to question. The intent of this paper is to examine the role of psychological variables in the explanation of social phenomena. This necessarily entails a delineation of the relationships between psychological and structural explanation. In doing this, I am in no way attempting to say the final word on the subject but merely to present what I think is a clarification useful in the formulation and analysis of problems. For my point of departure, I view explanation as involving two elements. The thing or event to be explained and the process of explanation itself. The event may be either positive or negative; i.e., we try to explain not only why certain events occur but also why certain ones do not occur. Analytically, the two approaches are quite different since one is based on observations in the real world and the other on logical considerations alone. For example, it is known that FZD marriage is associated with matrilineality. This is an empirical event of which we may inquire why it occurs and hope to find an explanation in human behavior. On the other hand, it is known that FZD marriage is not associated with matrilocality in matrilineal groups. It is again proper to inquire why this is so, but can we again hope to find an explanation in human behavior? I think that we might consider this possible, but I also believe that it is first necessary to inquire whether the association is logically possible. By this, I mean that it is acceptable to look for empirical explanations of what in fact exists, since anything that exists is obviously possible, but what does not exist is purely a construct and constructs must first be analyzed to determine whether they have logical validity. Thus, it may readily be seen